Ovarian cancer is the most lethal gynaecological malignancy in women [1
]. More than 70% of patients are diagnosed at an advanced stage as the disease lacks specific symptoms in early stages [1
]. The 5-year survival rate for women diagnosed at the late stage is less than 20%, whereas it is 90% if detected in the early stage [1
]. Therefore, early detection may result in better outcomes. CA125 is the most widely used serum marker in the detection and management of the disease [1
]. Normal CA125 concentrations are below 30 U/mL [2
]. CA125 levels are increased in 80%–85% of women in the advanced stages of ovarian cancer, as opposed to 50% of women in stage I of ovarian cancer. Therefore, CA125 is useful for monitoring progression and regression, rather than for early diagnosis [1
]. Elevated levels of CA125 have also been found elevated in benign conditions such as endometriosis, pregnancy, ovulatory cycles, liver diseases, congestive heart failure, and in infectious disease such as tuberculosis [2
]. CA125 is also one of several mucin cancer biomarkers. Other mucins, such as MUC1 (breast cancer) or prostate specific antigen (PSA, prostate cancer), have been used for cancer detection and progression monitoring, however, their sensitivity and specificity are limited. The quantities of these biomarkers in serum are low, as they are shed from tissue into the bloodstream [3
The quantification of soluble CA125 levels is currently performed with a second generation of assays. These assays are based on double determinant ELISA tests that use two monoclonal antibodies (mAb) directed against the M11 and OC125 epitope groups [2
]. Anti-CA125 antibodies are divided into three groups, OC125-like (group A), M11-like (group B), and Ov197-like which each recognize domains of non-overlapping epitopes [2
]. All three types of antibodies can recognize either native or denatured CA125. OC125 antibody exhibits a reduced binding after treatment with Peptide -N
-Glycosidase F (PNGase F) and currently available anti-CA125 antibodies do not permit fine discrimination among various CA125 species [2
CA125 antigen was first identified in six epithelial ovarian carcinoma cell lines and tumour tissue of ovarian cancer patients reacting with monoclonal antibody OC 125 [5
]. CA125 is a 200–2000 kDa mucin-type molecule (MUC16 mucin) with abundant N
- and O
-glycans (249 potential N
-glycosylation and over 3700 O
-glycosylation sites, Figure 1
). The carbohydrate content was previously estimated to be 24%–28% [6
]. It is a large transmembrane glycoprotein (Figure 1
]. The CA125 protein core is composed of a short cytoplasmic tail, a transmembrane domain and a large glycosylated extracellular structure [8
]. The extracellular domain is characterized by a large number of tandem repeats of SEA domains (sea-urchin sperm protein, enterokinase and agrin) which encompass an interactive disulfide bridged cysteine-loop and the site of OC125 and M11 binding [2
]. The molecule also includes an amino terminal domain of serine/threonine-rich sequences which may account for most of the O
-glycosylation known to be present in CA125 [8
]. Release or secretion of CA125 appears to be directly linked to the epithelial growth factor receptor signal transduction pathway [2
]. A functional role of CA125 in the physiological context or cancer remains unknown, however several properties of CA125 that may be of relevance for its biological function have been proposed. CA125 has been suggested to play a role as a lubricant, preventing adhesion of membranes [7
]. The glycan structures that are present on CA125 have been implicated in immune suppression, raising the possibility that CA125 might help protect the embryo from maternal immune rejection and play an immunoevasive role in ovarian cancer [7
CA125 binds to galectin-1 and mesothelin [2
]. CA125 from human peritoneal fluid was shown to enhance the invasiveness of a benign endometriotic cell line, which raises the possibility that CA125 plays a role in endometriosis [12
]. Several other mucins have been implicated in invasion and metastasis of cancer, partly because of similar functions. For example, MUC1 induces T cell apoptosis and increases invasiveness, MUC18 has been implicated in tumour angiogenesis, MUC2 enhances colon cancer metastasis to the liver, although it appears to inhibit initial neoplasia, MUC8 is up-regulated in metastatic medulloblastoma, and MUC3B is up-regulated in intestinal metaplasia [7
]. Previous studies investigated the glycosylation of CA125 from OVCAR3 cell lines, amniotic fluid and placenta, but not from serum. Wong et al.
analysed the major N
-glycans and O
-glycans of CA125 from the OVCAR3 cell line by mass spectrometry [9
]. In this study, 20% of the N
-glycans were found to be of the high mannose type and 80% of the complex type structures [9
]. Complex glycans were mostly mono-fucosylated bi-antennary, tri-antennary and tetra-antennary bisected structures with not more than one sialic acid [9
-glycans were both core 1 and 2 type glycans with branched core 1 antennae.
Milutinovic et al.
analysed CA125 glycans from amniotic fluid with lectins [13
]. VVA (Vicia villosa agglutinin, specific for terminal GalNAc) and SBA (Glycine max agglutinin, specific for GalNAcα1-Ser/Thr and/or GalNAcGalβ1,3GalNAcα1-Ser/Thr) showed the strongest binding [13
]. WGA (wheat germ agglutinin, specific for GlcNAc and its β1,4 oligomers) also reacted strongly, and LCA (Lens culinaris agglutinin, specific for Fucα1,6Man) and UEA (Ulex europaeus agglutinin, specific for Fucα1,2) showed lower reactivity [13
]. H2 antibody (recognising fucose α1,2 bound to terminal Gal epitopes) showed no reaction [13
]. The strong reaction of VVA and SBA indicates the predominance of O
-glycans and that the glycans are both core- and outer arm-fucosylated [13
]. Jankovic et al.
analysed CA125 from placenta with lectins [14
]. CA125 bound most strongly to WGA and RCA (Ricinus communis agglutinin, specific for terminal Gal linked β1,4 to GlcNAc), but low affinity interactions occurred with the other lectins such as ConA (Canavalia ensiformis, specific for high mannose glycans), SNA (Sambucus nigra agglutinin, specific for sialic acid linked α2,6 to Gal), MAA (Maackia amurensis agglutinin, specific for sialic acid linked α2,3 to Gal), AAA (Aleuria aurantia agglutinin, specific for core-fucosylated N
-glycans, outer arm-fucosylated glycans are retarded), SBA and PNA (Arachis hypogaea agglutinin, specific for Galβ1,3 GalNAcα1-Ser/Thr) [14
]. Placental CA125 showed both core- and outer arm-fucosylated, mono- and di-sialylated glycans linked both α2,3 and α2,6 [14
]. WGA and RCA reactivity indicated the presence of polylactosamine structures or the presence of O
-glycans. Based on PNA and SBA binding, these O
-glycans are Galβ1–3GalNAc and (Galβ1–3GlcNAcβ1–6) GalNAc [14
In our previous glycosylation studies on breast cancer MUC1, pancreatic cancer RNAse1 and prostate cancer PSA, we found different glycosylation in tumour origins [3
]. In this study we attempted to investigate the glycosylation of CA125 as a possible improved serum biomarker for the detection of ovarian cancer. We used controls with CA125 values above normal 30 U/mL to evaluate significance of our results.
4. Experimental Section
4.1. Serum Samples
Patient serum samples were obtained from St. Vincent’s University Hospital Dublin. These patients had CA125 requested and all had advanced ovarian cancer. All were Caucasian between 45 and 70 years of age, their CA125 levels were 215.8–1977 U/mL. Following measurement of CA125, all samples were anonymised and used for glycan analysis. Control sera obtained from Innovative research (Patricell Ltd) were from 58 and 61 years old white females, their CA125 levels were 63.2 U/mL and 39.8 U/mL. Samples were collected with informed consent from healthy donors. 450 mL of whole blood was collected on a cold pack, centrifuged at 5000× g for 10 min. After centrifugation, the serum was transferred from the red blood cells to a blood transfer bag. The serum was stored at room temperature for 24–48 h until it had finished clotting. After the serum had clotted, it was centrifuged ant 5000× g for 15 min. The serum was transferred to a sterile plastic bottle and stored at −20 °C.
4.2. CA125 Isolation from Serum Samples
Method of immunoadsorption according Peter et al.
for PSA [17
] was optimized for CA125 isolation. Four milligrams of streptavidin coated magnetic beads (Roche, 10 mg/mL) were washed three times with 400 μL of washing buffer (10 mM TRIS, 150 mM NaCl, pH 7.5, 0.5% Tween 20) using magnetic separation. Meantime, 60 μg biotinylated mouse monoclonal antibody anti-CA125 (Hytest, MAb X306, 2.9 mg/mL) was prepared in 400 μL of incubation buffer (10 mM TRIS, 150 mM NaCl, pH 7.5, 0.1% Tween-20, 1% BSA) and added to the washed beads. The mixture was incubated for 30 min at room temperature (RT) under gentle shaking. The beads were then washed three times with 400 μL of washing buffer. 400 μL of serum was prepared (containing approximately 300 U = 1.2 μg of CA125) and added to the washed beads with bound anti-CA125 antibody. The mixture was incubated for 1 h, at RT, slightly shaking. The beads were then washed three times with 400 μL of washing buffer. CA125 was eluted with 250 μL of water for 1 h at RT, slightly shaking and the beads were washed with 50 μL of water; and then eluted with 250 μL of formic acid/water/acetonitrile (1:3:2) for 1 h, RT, slightly shaking and the beads were washed with 50 μL of formic acid/water/acetonitrile (1:3:2) and added to the eluate.
4.3. SDS-PAGE Electrophoresis
Electrophoresis in 4%–12% Bis-Tris SDS-PAGE mini-gels (Invitrogen, Carlsbad, CA, USA) was performed at room temperature according to the method of Laemmli [29
]. The gels were Coomassie and Ponceau stained. All samples were reduced with 5% 2-mercaptoethanol before analysis.
4.4. Western Blot
Isolated proteins from samples were after electrophoresis transferred to a PVDF membranes (400 mA constant for 2 h) and Ponceau stained. Membranes were blocked with Tris-buffered saline Tween (TBST, 10 mM TRIS, 100 mM NaCl, pH 7.5, 0.2% Tween 20) with 5% dry milk for 1 h at RT. Membranes were washed three times 10 min with TBST before an overnight incubation at 4 °C in 2.9 μg/mL anti-CA125 (Hytest, MAb X306) in the TBST-1% dry milk. Membranes were washed three times 10 min with TBST before 1.5 h incubation with 0.04 μg/mL secondary rabbit anti-mouse (Abcam, Ab6728). The blots were developed using the ECL Plus chemiluminescent detection system (GE Healthcare, Uppsala, Sweden).
4.5. Release and Purification of N- and O-Glycans from Human Serum CA125 in Gel Block
-glycans were released from serum CA125 by in situ
digestion with Peptide -N
-Glycosidase F (PNGase F, Roche, Mannheim, Germany) in SDS-PAGE gel bands as described earlier [28
]. Briefly, isolated CA125 gel bands obtained from SDS-PAGE were alkylated, washed and N
-glycans released by PNGase F. O
-glycans were released from SDS-gel blocks after N
-glycans release using ammonia-based β-elimination [30
4.6. Fluorescent Labelling of the Reducing Terminus of N-Glycans
Glycans were fluorescently labelled with 2-aminobenzamide (2AB) by reductive amination [31
] (LudgerTag 2-AB labelling kit LudgerLtd., Abingdon, UK).
4.7. Exoglycosidase Digestion of 2AB Labelled N-Linked Glycans
All enzymes were purchased from Prozyme, San Leandro, CA, USA. The 2AB-labelled glycans were digested in a volume of 10 μL for 18 h at 37 °C in 50 mM sodium acetate buffer, pH 5.5, using arrays of the following enzymes: ABS—Arthrobacter ureafaciens sialidase (EC 22.214.171.124), NAN1—Streptococcus pneumoniae sialidase (EC 126.96.36.199), 1 U/mL; BTG—bovine testes β-galactosidase (EC 188.8.131.52), 1 U/mL; BKF—bovine kidney alpha-fucosidase (EC 184.108.40.206), 1 U/mL; GUH—β-N-acetylglucosaminidase cloned from Streptococcus pneumonia, expressed in E. coli (EC 220.127.116.11), 4 U/mL.
After incubation, enzymes were removed by filtration through a protein binding EZ filters (Millipore Corporation, Beford, MA, USA) [32
], the N
-glycans were then analysed by HILIC.
4.8. Hydrophilic Interaction Liquid Chromatography (HILIC)
HILIC was performed using a TSK-Gel Amide-80 4.6 × 250 mm column (Anachem, Luton, UK) on a 2695 Alliance separations module (Waters, Milford, MA, USA) equipped with a Waters temperature control module and a Waters 2475 fluorescence detector. Solvent A was 50 mM formic acid adjusted to pH 4.4 with ammonia solution. Solvent B was acetonitrile. The column temperature was set to 30 °C. The 3 h gradient started with a linear gradient of 20%A and went up continuously over 152 min to 58%A at a flow rate of 0.4 mL/min. Samples were injected in 80% acetonitrile [18
]. Fluorescence was measured at 420 nm with excitation at 330 nm. The system was calibrated using an external standard of hydrolyzed and 2AB-labelled glucose oligomers to create a dextran ladder, as described previously [32
]. Experimentally determined reproducibility for the quantitation of the HILIC peaks was found to be 2%–30% (in average 9%) relative standard deviation using ten individually prepared and analysed aliquots of serum from the same sample (Saldova et al.
, in preparation).
4.9. Weak Anion Exchange Chromatography (WAX)—High Performance Liquid Chromatography (HPLC)
WAX-HPLC was performed using a Vydac 301VHP575 7.5 × 50-mm column (Anachem) on a 2695 Alliance separations module with a 474 fluorescence detector (Waters). Solvent A was 0.5 M formic acid adjusted to pH 9.0 with ammonia solution, and solvent B was 10% (v
) methanol in water. Gradient conditions were as follows: a linear gradient of 0 to 5% A over 12 min at a flow rate of 1 mL/min, followed by 5% to 21% A over 13 min and then 21% to 50% A over 25 min, 80% to 100% A over 5 min, and then 5 min at 100% A. Samples were injected in water. A fetuin N
-glycan standards were used for calibration [32
4.10. Negative Ion Electrospray Ionisation Mass Spectrometry ESI-MS and ESI MS/MS
Samples were analysed by static nanoelectrospray ionization using a Waters (Waters MS Technologies, Manchester UK) tandem quadrupole time-of-flight mass spectrometer. Samples were diluted with a 1:1 (v:v
) mixture of water:methanol containing 0.1 mM ammonium phosphate. MS and MS/MS data was acquired in negative mode with the following instrument settings: source temperature 120 °C, capillary voltage 1.3 kV, cone voltage 100 V and the RF-1 voltage was 130 V. MS/MS precursor ions were selected with a 3 m/z mass window and were fragmented by CID using argon as the collision gas. CID voltage was altered accordingly from 20 to 40 V. Data acquisition and data processing were conducted with Waters MassLynx version 4.1. Interpretation of the negative ion MS/MS spectra was according to published work [34
4.11. Protein Identification by Mass Spectrometry (MS)
Samples were run on a Thermo Scientific LTQ ORBITRAP XL mass spectrometer connected to an Exigent NANO LC.1DPLUS chromatography system. Tryptic peptides were resuspended in 0.1% formic acid. Each sample was loaded onto a Biobasic C18 PicofritTM column (100 mm length, 75 mm ID) and was separated by an increasing acetonitrile gradient, using a 60 min reversed phase gradient (7%–40% acetonitrile for 40 min) at a flow rate of 300 nL/min. The mass spectrometer was operated in positive ion mode with a capillary temperature of 200 °C, a capillary voltage of 46 V, a tube lens voltage of 140 V and with a potential of 1900 V applied to the frit. All data was acquired with the mass spectrometer operating in automatic data data-dependent switching mode. A high resolution MS scan was performed using the Orbitrap to select the 5 most intense ions prior to MS/MS analysis using the Ion trap. The precursor accurate mass was <20 ppm and the MS/MS fragment mass tolerance was + or −0.8 Da.
The raw data was analysed using Bioworks Browser 3.3.1 SP1, a proteomics analysis platform. All MS/MS spectra were sequence database searched using the algorithim TurboSEQUEST. The MS/MS spectra were searched against a non-redundant human Swissprot database. The following search parameters were used: precursor-ion mass tolerance of 20 ppm, fragment ion tolerance of 1.0 Da with methionine oxidation and cysteine carboxyamidomethylation specified as differential modifications and a maximum of 2 missed cleavage sites allowed. Each peptide used for protein identification met specific parameters, i.e., XCorr values of ≥1.9, ≥2.5, ≥3.2 for single-, double-, and triple- charged peptides, respectively, and a peptide probability of <0.001.
4.12. CA125 ELISA
CA125 levels after immunoadsorption were measured using Cancer antigen CA125 enzyme immunoassay test kit (BC-1013, BioCheck, Foster City, CA, USA) according to manufacturer’s instructions.